train_utils.py 7.61 KB
Newer Older
1
2
3
4
5
6
7
8
9
# @author lucasmiranda42
# encoding: utf-8
# module deepof

"""

Simple utility functions used in deepof example scripts. These are not part of the main package

"""
10
from datetime import date, datetime
11

12
from kerastuner import BayesianOptimization, Hyperband
13
from kerastuner import HyperParameters
14
from kerastuner_tensorboard_logger import TensorBoardLogger
15
from typing import Tuple, Union, Any, List
16
17
18
19
20
21
22
import deepof.hypermodels
import deepof.model_utils
import numpy as np
import os
import pickle
import tensorflow as tf

23
24
hp = HyperParameters()

25

26
def load_hparams(hparams):
27
28
29
30
31
32
33
34
35
36
37
38
    """Loads hyperparameters from a custom dictionary pickled on disc.
    Thought to be used with the output of hyperparameter_tuning.py"""

    if hparams is not None:
        with open(hparams, "rb") as handle:
            hparams = pickle.load(handle)
    else:
        hparams = {
            "units_conv": 256,
            "units_lstm": 256,
            "units_dense2": 64,
            "dropout_rate": 0.25,
39
            "encoding": 16,
40
41
42
43
44
45
46
47
48
49
50
51
            "learning_rate": 1e-3,
        }
    return hparams


def load_treatments(train_path):
    """Loads a dictionary containing the treatments per individual,
    to be loaded as metadata in the coordinates class"""
    try:
        with open(
            os.path.join(
                train_path,
52
                [i for i in os.listdir(train_path) if i.endswith(".pkl")][0],
53
54
55
56
57
58
59
60
61
62
63
            ),
            "rb",
        ) as handle:
            treatment_dict = pickle.load(handle)
    except IndexError:
        treatment_dict = None

    return treatment_dict


def get_callbacks(
lucas_miranda's avatar
lucas_miranda committed
64
65
66
67
68
69
    X_train: np.array,
    batch_size: int,
    cp: bool,
    variational: bool,
    predictor: float,
    loss: str,
70
) -> List[Union[Any]]:
71
    """Generates callbacks for model training, including:
72
73
74
75
    - run_ID: run name, with coarse parameter details;
    - tensorboard_callback: for real-time visualization;
    - cp_callback: for checkpoint saving,
    - onecycle: for learning rate scheduling"""
76

77
    run_ID = "{}{}{}_{}".format(
78
79
80
81
82
83
84
        ("GMVAE" if variational else "AE"),
        ("P" if predictor > 0 and variational else ""),
        ("_loss={}".format(loss) if variational else ""),
        datetime.now().strftime("%Y%m%d-%H%M%S"),
    )

    log_dir = os.path.abspath("logs/fit/{}".format(run_ID))
85
    tensorboard_callback = tf.keras.callbacks.TensorBoard(
86
87
88
        log_dir=log_dir,
        histogram_freq=1,
        profile_batch=2,
89
90
91
    )

    onecycle = deepof.model_utils.one_cycle_scheduler(
92
93
        X_train.shape[0] // batch_size * 250,
        max_rate=0.005,
94
95
    )

96
97
98
99
100
101
102
103
104
105
106
107
108
    callbacks = [run_ID, tensorboard_callback, onecycle]

    if cp:
        cp_callback = tf.keras.callbacks.ModelCheckpoint(
            "./logs/checkpoints/" + run_ID + "/cp-{epoch:04d}.ckpt",
            verbose=1,
            save_best_only=False,
            save_weights_only=True,
            save_freq="epoch",
        )
        callbacks.append(cp_callback)

    return callbacks
109
110
111


def tune_search(
112
    data: List[np.array],
113
114
    hypertun_trials: int,
    hpt_type: str,
115
116
    hypermodel: str,
    k: int,
117
    kl_warmup_epochs: int,
118
    loss: str,
119
    mmd_warmup_epochs: int,
120
    overlap_loss: float,
121
    pheno_class: float,
122
123
    predictor: float,
    project_name: str,
124
    callbacks: List,
125
    n_epochs: int = 30,
126
    n_replicas: int = 1,
127
) -> Union[bool, Tuple[Any, Any]]:
128
129
    """Define the search space using keras-tuner and bayesian optimization

130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
    Parameters:
        - train (np.array): dataset to train the model on
        - test (np.array): dataset to validate the model on
        - hypertun_trials (int): number of Bayesian optimization iterations to run
        - hpt_type (str): specify one of Bayesian Optimization (bayopt) and Hyperband (hyperband)
        - hypermodel (str): hypermodel to load. Must be one of S2SAE (plain autoencoder)
        or S2SGMVAE (Gaussian Mixture Variational autoencoder).
        - k (int) number of components of the Gaussian Mixture
        - loss (str): one of [ELBO, MMD, ELBO+MMD]
        - overlap_loss (float): assigns as weight to an extra loss term which
        penalizes overlap between GM components
        - pheno_class (float): adds an extra regularizing neural network to the model,
        which tries to predict the phenotype of the animal from which the sequence comes
        - predictor (float): adds an extra regularizing neural network to the model,
        which tries to predict the next frame from the current one
        - project_name (str): ID of the current run
        - callbacks (list): list of callbacks for the training loop
        - n_epochs (int): optional. Number of epochs to train each run for
        - n_replicas (int): optional. Number of replicas per parameter set. Higher values
         will yield more robust results, but will affect performance severely

    Returns:
        - best_hparams (dict): dictionary with the best retrieved hyperparameters
        - best_run (tf.keras.Model): trained instance of the best model found
154
155
156

    """

157
158
    X_train, y_train, X_val, y_val = data

159
160
161
162
    assert hpt_type in ["bayopt", "hyperband"], (
        "Invalid hyperparameter tuning framework. " "Select one of bayopt and hyperband"
    )

lucas_miranda's avatar
lucas_miranda committed
163
    if hypermodel == "S2SAE":  # pragma: no cover
164
165
166
        assert (
            predictor == 0.0 and pheno_class == 0.0
        ), "Prediction branches are only available for variational models. See documentation for more details"
167
        hypermodel = deepof.hypermodels.SEQ_2_SEQ_AE(input_shape=X_train.shape)
168
169
170

    elif hypermodel == "S2SGMVAE":
        hypermodel = deepof.hypermodels.SEQ_2_SEQ_GMVAE(
171
            input_shape=X_train.shape,
172
            kl_warmup_epochs=kl_warmup_epochs,
173
            loss=loss,
174
            mmd_warmup_epochs=mmd_warmup_epochs,
175
            number_of_components=k,
176
            overlap_loss=overlap_loss,
177
            phenotype_predictor=pheno_class,
178
            predictor=predictor,
179
        )
lucas_miranda's avatar
lucas_miranda committed
180

181
182
183
    else:
        return False

184
185
186
187
188
189
190
191
192
193
194
195
196
197
    hpt_params = {
        "hypermodel": hypermodel,
        "executions_per_trial": n_replicas,
        "logger": TensorBoardLogger(metrics=["val_mae"], logdir="./logs/hparams"),
        "objective": "val_mae",
        "project_name": project_name,
        "seed": 42,
        "tune_new_entries": True,
    }

    if hpt_type == "hyperband":
        tuner = Hyperband(
            directory="HyperBandx_{}_{}".format(loss, str(date.today())),
            max_epochs=hypertun_trials,
198
            hyperband_iterations=3,
lucas_miranda's avatar
lucas_miranda committed
199
            factor=2,
200
201
202
203
204
205
206
207
            **hpt_params
        )
    else:
        tuner = BayesianOptimization(
            directory="BayOpt_{}_{}".format(loss, str(date.today())),
            max_trials=hypertun_trials,
            **hpt_params
        )
208
209
210

    print(tuner.search_space_summary())

211
212
213
214
215
216
217
218
219
220
221
    Xs, ys = [X_train], [X_train]
    Xvals, yvals = [X_val], [X_val]

    if predictor > 0.0:
        Xs, ys = X_train[:-1], [X_train[:-1], X_train[1:]]
        Xvals, yvals = X_val[:-1], [X_val[:-1], X_val[1:]]

    if pheno_class > 0.0:
        ys += [y_train]
        yvals += [y_val]

222
    tuner.search(
223
224
        Xs,
        ys,
225
        epochs=n_epochs,
226
        validation_data=(Xvals, yvals),
227
228
        verbose=1,
        batch_size=256,
lucas_miranda's avatar
lucas_miranda committed
229
        callbacks=callbacks,
230
231
232
233
234
    )

    best_hparams = tuner.get_best_hyperparameters(num_trials=1)[0]
    best_run = tuner.hypermodel.build(best_hparams)

lucas_miranda's avatar
lucas_miranda committed
235
236
    print(tuner.results_summary())

237
238
239
240
241
242
243
244
    return best_hparams, best_run


# TODO:
#    - load_treatments should be part of the main data module. If available in the main directory,
#    a table (preferrable in csv) should be loaded as metadata of the coordinates automatically.
#    This becomes particularly important por the supervised models that include phenotype classification
#    alongside the encoding.